Arsenic in Drinking Water Toxicological Risk Assessment in the North Region of Burkina Faso

Journal of Water Resource and Protection, 2013, 5, 46-52 http://dx.doi.org/10.4236/jwarp.2013.58A007 Published Online August 2013 (http://www.scirp.org/journal/jwarp)

Jean Fidèle Nzihou1*, Médard Bouda2, Salou Hamidou3, Jean Diarra4 1École Normale Supérieure/Université de Koudougou, Koudougou, Burkina Faso 2Institut de Recherche en Mine et Environnement, Université de Québec en Abitibi-Témiscamingue, Rouyn-Noranda, Canada 3Université Catholique de l’Afrique de l’Ouest (UCAO), Unité Universitaire à Bobo-Dioulasso, Bobo-Dioulasso, Burkina Faso 4Institut de Génie de l’Environnement et du Développement Durable/Université de Ouagadougou, Ouagadougou, Burkina Faso Email: *[email protected]

Received February 4, 2013; revised March 6, 2013; accepted April 2, 2013

Copyright © 2013 Jean Fidèle Nzihou et al. This is an open access article distributed under the Creative Commons Attribution Li- cense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

ABSTRACT Human health risks assessment were estimated by determining the nature and probability of adverse health effects in the North region’s populations who are now exposed to arsenic from drinking water or will be exposed in the future. Sev- eral questions were addressed in this study: what types of health problems may be caused by arsenic from drinking water? What is the chance that people will experience health problems when exposed to different levels of arsenic? What arsenic level are people exposed to and for how long? To answers these questions we have first identified the hazard by evaluating arsenic concentration in thirty-four (34) bore-hole water points among the region based on the assumption of clinical cases related to drinking water. Arsenic concentration ranged from 0 up to 87.8 micrograms per liter. Next we assessed the dose-response of exposure to arsenic. Dose-response relationship describes how the likelihood and severity of adverse health effects are related to the amount and condition of exposure to arsenic. This required us to choose toxicity reference values (TRVs) above which adverse effects may occur for noncarcinogenic and for carcinogenic effects. Exposure factors have been calculated in two scenarios: people from 0 to 14 years old and people from 15 to 70 years. Exposure has been estimated indirectly through consideration of measured concentrations of arsenic in drinking water. This study show that people in the Yatenga, Zondoma and Passore provinces are at very high risk for developing several pathologies such as hyper pigmentation, keratosis, cancer, etc. due by chronic exposure to arsenic in drinking water.

Keywords: Arsenic Poisoning; Toxicity Reference Value; Risk Assessment; Carcinogenic Effect; Exposure Factor

1. Introduction this obvious problem of water scarcity in this region, there is also another due to remarkable pollution of ground- The North region, one of the thirteen (13) of Burkina water. The rainfall is insufficient and erratic with annual Faso, has many economic and cultural potentialities. But precipitation of around 600 to 700 mm. Banks dams are it faces an acute lack of drinking water. Indeed, access to operated seasonally for market gardening activities and drinking water poses a major problem to the population rivers are increasingly solicited by gold mining, mercury both for its quality and its management. All water sources and cyanide are even used in the artisanal extraction of are exploited in order to meet the needs of drinking water. gold and this in the beds of rivers. There is a growing This region had respectively 51,595,875 and 6292 mod- concern about the scale of anthropogenic pollution in this ern water access points but only 9,401,074 and 1129 of region. these were permanents respectively for years 2007, 2009 In these fifteen (15) years, the appearance of clinical and 2011 [1]. It appears that for drilling and large di- cases of hyperkeratosis, hyper pigmentation and ne- ameter wells, with a population of 1,185,796 inhabi- crotizing ulcerative tumor in some villages in the North tants (2006 census), here is an average of 205 people by region have attracted the attention of public authorities a modern water access point. Yet this estimation doesn’t on the issue of diseases from water. Therefore the dif- take in account the grown of the population. Apart from ferent structures of the ministries of water, health, envi- *Corresponding author. ronment and development partners (DANIDA, UNICEF,

WHO) are highly involved in resolving this problem. The work reported in reference [2] has revealed the presence of arsenic in 159 tube-wells water at concentrations above the current standard of 50 μg·L−1 in Burkina Faso. Groundwater contamination by arsenic levels is up to 1630 μg·L−1 due to mining activities having been measured [3]. The analysis and epidemiological study reported in reference [4] has established a high correlation between the presence of Arsenic in patients’ urines and the concentration of Arsenic in their source of drinking water in Yatenga, one of the four provinces of the North region. Solution to this problem was supposed to drill 100 new bore-hole in order to replace access point above national standards. But reference [1] indicates that only 55 new permanents water access points were created Figure 1. Delimitation of the North region. between 2009 and 2011. Maybe the situation has wors- ened in the meanwhile. After the newsworthy effect due from September 2006 to December 2006. The sampling to death caused by Arsenic contaminated water, Arsenic period may seem short and old. But it was shown that probably continues to silently endanger the health and “The arsenic content in water decreases with the produc- kill people in the North region of Burkina Faso. The tivity of drilling, but arsenic content in water content above facts raised the following questions: change from one site to another” [6]. Conclusion was  What are the levels populations’ exposure to arsenic? also made that “it is the geological context which the  From drinking water in the North region of Burkina source of arsenic level in groundwater”. Besides “Arse- Faso? nic contamination of groundwater has been found to oc-  What are the effects of arsenic on human health? cur due to geothermal influence to groundwater, mineral What health risk does undertake population when it is dissolution (e.g., pyrite oxidation), desorption in the oxi- not possible to close an arsenic contaminated water ac- dizing environment and reductive desorption and disso- cess point? lution” [7,8]. The present study addresses these issues through a Several sampling points were selected based on the quantitative heal risk assessment based on data collected following criteria: in the region. The study aims to predict in the medium  Proximity to risk areas based on the assumption of and long run what health risk does encounter populations clinical cases; of this region exposed to arsenic in drinking water.  Location and clear name (no ambiguity) of the sampling points; 2. Materials and Methods  Easy access to the sampling points. Each sample was labeled with the following informa- 2.1. Study Area tion: the name of the operator, identification number of The North Region is located in the northern part of the sample, sampling point location, sampling date and Burkina Faso and extend between latitudes 12˚38' and nature (surface water or groundwater). Physical parame- 14˚18' North and longitude 1˚33' and 2˚55' West [5]. The ters such as, alkalinity, electrical conductivity (Eh), dis- region is bordered by the republic of Mali at the north solved oxygen and temperature were measure din situ and five other regions of Burkina Faso, namely: Sahel, (during sampling) to avoid their alteration, even well Central North, Central Plateau, Central West and Mou- preserved before arriving at the laboratory. houn respectively at the north-east, east, south and west Laboratory analysis of the samples was done using: as depicted in Figure 1.  An automatic absorption spectrometer (AAS) graphite furnace for the determination of aluminum, cad- 2.2. Sampling mium, chromium, nickel and lead. Automatic dosing of the sample is done in the graphite AAS furnace. Samples were collected from the four (4) provinces of Thereafter there is a direct determination of the con- the North region. Random sampling was used and no rule centration of each element using a calibration curve, has been followed in relation to the space distribution of or from the specific absorbance of each element using the sampled points. Thus, the number of samples per the spectrometer is equipped with a background cor- province does not depend on the size of the province, but rection system continuously after performing a cor- the presumption of clinical cases. Samples were collected rection to the non-specific absorbance.

 A FIAS (Flow Injection for Atomic Spectroscopy) Among all these TRVs, we have taken the sixth, which system for the determination of arsenic, mercury and is 7×10−5 mg·kg−1·d−1. For selecting the TRV we used the selenium. Before the analysis of mercury, the subset study reported in reference [9]. TRV selection criteria of sample is digested with nitric acid under the pres- from this study are: sure of one bar. And, for the analysis of arsenic, the  TRV obtained from the most recent study know to us. subset of sample is reduced with chlorine acid (HCl),  Study conducted over a long period of 6 months to ascorbic acid and potassium iodine in an oven. The 33 years for arsenic concentrations <10 mg·L−1 and −1 test sample is introduced into the feedback loop and >229 mg·L . automatically mixed with HCl and NaBH4. The reac-  The exposure route is oral. tion produces a volatile hydride that is transported  The study highlights the critical effects such as skin into the quartz cell using a carrier gas as argon. In the lesions (hyper pigmentation, hypo pigmentation, ke- cell hydride volatile is converted to gaseous metal ratosis) which represent the main non-carcinogenic atoms. The determination of the total concentration of effects due to chronic oral exposure to arsenic. calcium and magnesium and the total hardness is ma-  The method developed in this study is a multi-stages de by mole titration of calcium and magnesium ions logistic regression to estimate odds ratios for skin le- with a solution of hydroxy naphtol and ethylenedia- sions based on different levels of exposure to arsenic. mintetraacetic acid (EDTA) at pH 10. Black Tri Factors that may influence the multi-stages regression chrome which gives a dark red or purple in the pres- of more than 5% were included in the model (age, sex, ence of calcium and magnesium ions, is used as an education) and level of housing. They obtained a sig- indicator. nificant dose-response relationship between cumula- The analysis in this study targeted a large number of tive exposure or average arsenic and the risk of de- parameters: temperature, pH, conductivity, turbidity, the veloping skin lesions has been demonstrated, the low- alkaline title, the fluoride ion, sulfates, phosphate, the est category of exposure was used as control. sodium ions, potassium (K), calcium (Ca), magnesium  The TRV is one of that best protects the population. (Mg), etc. Temperature measured in situ showed no sig- For stochastic effects: several TRVs were determined nificant variation and was about 27˚C ± 1˚C and will not by various agencies including the US EPA, Health be reported in this paper. Canada and the OEHHA. Table 2 summarizes these values. 2.3. Toxicity Reference Values (TRVs) TRVs currently available are those related to skin cancer by the oral exposure. We should keep in mind that The effects studied are the deterministic and stochastic the carcinogenic effects of arsenic are also linked to bla- effects related to drinking water. dder cancer and lung cancer [10]. For deterministic effects: there are several TRVs for We choose the value of the US EPA (1.5 mg·kg−1·d−1) arsenic available in the literature. These are summarized which is the same as that proposed by OEHHA. Both are in Table 1 below. the most recent. This value was determined from the syn- thesis report made by US EPA in 1988. Studies used to Table 1. Available TRVs for threshold effects. establish the dose-response relation are those in refer- Exposure Uncertainty ences [11,12]. The critical effect retained is skin cancer. Sourcea Reference valueb Revision year route factor A multi-stages linear and quadratic model based on the RfD = 3 × 10−4 US EPA Oral 3 1993 prediction of apparition of cutaneous cancer according to mg/kg/d dose and age was used. MRL = 5 × 10−3 ATSDR Oral (acute) 10 2000 mg/kg/d MRL = 3 × 10−4 3. Results and Discussion ATSDR Oral 3 2005 mg/kg/d TDI = 10−3 Thirty-four (34) collected samples were analyzed. These RIVM Oral 2 2001 mg/kg/d samples (Table 3) are distributed as follow: 16 (55.88%) REL = 3 × 10−4 OEHHA Oral 3 2005 in the Yatenga province, 09 in Passore (26.47%), 5 in mg/kg/d Zondoma (11.76%) and 4 in Loroum (11.76%). VTR = 7 × 10−5 INERIS Oral 10 2007 mg/kg/d 3.1. Exposures Estimation aUS EPA = United States Environmental Protection Agency, ATSDR = Agency for Toxic Substances and Diseases Registry, RIVM = “RijskInsti- 3.1.1. Determination of Emissions tuutvoor Volksgezond heid en Milieu” (National Institute of Public Health and Environment in Netherands), OEHHA = Office of Environmental According to the 2006 census [13] the North region had Health Hazard Assessment, INERIS = Institut National de l’Environnement 1,185,796 inhabitants divided as follow: 553,164 in Insustriel et des RISques. bRfD = Refernce Dose, MRL = Maximum Risk Level, TDI = Tolerable Daily Intake, VTR = Valeur Toxicologique de Ré- Yatenga, 323,222 in Passoré, 166,557 in Zondoma and férence (Toxicity Reference Value). 142,853 in Loroum. The population of the study area

Table 2. Available VTRs for chronic oral exposure to arse- habitants, Passoré 265,947 inhabitants, Zondoma 137,043 nic for non-threshold effects. inhabitants and Loroum 117,539 inhabitants . Chemical Exposure Reference Revision Source substance route valuea year 3.1.2. Population Exposure Inorganic REL = 1.5 The major exposure route to arsenic in these populations US EPA Oral 1998 Arsenic (mg/kg/d)−1 is bore-wells water and more specifically drinking water Inorganic Santé TD = 1.8 × from boreholes. Exposure duration range from 6 months Oral 0.05 2004 Arsenic Canada 10−2 mg/kg/d to 33 years. Exposure levels of populations to arsenic −1 −1 Inorganic REL = 1.5 ranged from less than 10 μg·L to over 87 μg·L . We OEHHA Oral 2005 Arsenic (mg/kg/d)−1 consider an exposure duration of 33 years in our sce- a REL = Reference Exposure Level, TD0.05 = Tumoric Dose 0.05. nario.

Table 3. pH, K, Ca and As in collected samples values. 3.1.3. Calculation of the Exposures K Ca As We are here in the context of an oral contamination. Ex- Departement Sampling place pH (µg/L) (µg/L) (µg/L) posure dose (ED) and exposure factor (EF) can be calcu- Koeyiri 7.1 4.1 25.65 1 lated with the following equations [14]: Arbole CSPS 6.4 2.7 16.03 1 ED C IR AF EF BW (1) F. CSPS Sarma 7.13 1.7 68.94 87.8 Bokin F. Ecole Rana 7.2 1.1 55.31 10.8 The exposure factor express how often and how long a Gourcy F. Yipala Koenba 7.1 4.2 19.24 8.9 person may be contacting a toxicant in environment. Exposure Factor = ED × (F/AT), that is: Ramiga 6.6 0.5 24.05 1 Kalsaka Zingdegin 7.5 0.7 17.64 1 ED C IR AF F BW  AT  (2) Bissighin-Saabin 7.5 2.5 48.9 16.8 where: Kindibo F. Silmi-mosse 7.3 4.7 36.07 6.3 C = arsenic concentration in water (milligram//liter) Nabadogo 7.2 3.7 28.06 5.8 Kossouka IR = Intake rate (kg/day) Deira 7.2 0.9 63.33 1.1 AF = Bioavailability factor (unitless) Pougyango 6.6 0.5 22.4 1.6 Passoré EF = Exposure factor (unitless) Lemessere 6.8 3.4 24.05 1.5 BW= Body weight (kilograms) Ramgouma/Itaore 7.35 5.5 42.48 5.6 F = Frequency of exposure (days/year) Nayiri 7.3 1.7 50.5 9 Ed = Exposure duration (years) Rambo Ukoudin 7 4.4 27.25 1 AT = Averaging time (Ed × 365 days/year) Massogo-yiri 7 1.8 39.28 1 Exposure factor have been calculated for non-cancer Tiibi-Zinczougou 7 0.5 21.64 1 effects and carcinogenic effects. Paramyiri 6.85 5.8 56.11 1 3.1.4. Non-Carcinogenic Effects Secteur 2 Toghin 7.55 0.5 52.91 1 Seguenega Exposure factor to arsenic in drinking water was calcu- Tingandin 6.5 2.5 19.24 1 lated according to two scenarios: the first for “0 - 14 Tognayiri 7.15 3.1 28.06 1 years” and the second “15 - 70 years”. Tangaye Bonsomnomogo 7.5 0.5 51.3 1.2 In cases of chronic arsenic poisoning, skin lesions pre- CSPS 7.15 1.2 16.03 2.1 dominate: hyperkeratosis of the palms and soles, dotted Titao Ecole/Noogo 7.35 8.2 89.79 1 hyper pigmentation concomitant with small hypo pig- F. DPAHRH 6.95 4.2 133 1 mented areas. In cases of poisoning by drinking water, F. CSPS Rikba 6.95 4.5 23.25 8.5 these cutaneous signs are the most sensitive effects of Touogo F. Bankpore Ranawa 7.1 0.5 16.03 6.5 arsenic on humans. Keratoses appear to be the earliest F. Ecole Kelgaum 7.5 1.3 16.03 15.1 signs of arsenic poisoning. Forage de kouma 7.6 1.5 32.06 7.4 Exposure doses calculated in Tables 4 and 5 are far higher than that of the maximum daily intake in Yatenga, Rosin 7.2 0.5 57.72 9 Yako Passoré and Zondoma provinces. This is illustrated by Itaore 7.1 1.9 56.11 1.9 clinical cases of hyper pigmentation seen in the region (Figure 2). subject to contamination by arsenic includes all residents (men, women, children older than 4 years), living and 3.1.5. Carcinogenic Effects drinking water in the region. There are 975,673 individu- Major cancers associated with arsenic exposure are can- als, divided by province as follow: Yatenga 455,143 in- cers of the skin, bladder, lung, kidney and liver. Table 6

Table 4. Exposure doses for non-carcinogenic effects with intake of the chemical over a specific time period with scenario 1. the RfD for that chemical derived for a similar period of Scenario 1: Exposed population: individuals from 0 to 14 years exposure. This comparison results in a non-carcinogenic Ed: 7 years, IR (kg/d) = 1, BW (kg) = 29, Hazard Quotient (HQ), as follows: Ed (years) = 7, Total Period (years) = 7 HQ DI RfD (3) Province Yatenga Passoré Zondoma Lorom where: C (µg/L) 10 16 10 87 10 15 0 0 HQ = Hazard Quotient −1 −1 ED (μg·kg ·d ) 0.35 0.55 0.35 3 0.35 0.52 0 0 DI = Daily Intake (mg/kg-day) RfD = Reference Dose (mg/kg-day) −5 −1 −1 Table 5. Exposure doses for non-carcinogenic effects with with a RfD of 7×10 mg·kg ·d , calculated HQ for the scenario 2. North region of Burkina Faso are summarized in Table 7. Scénario 1: Exposed population: individuals from 15 to 65 years In the second scenario, individuals from age 15 to 70 Ed: 33 years, IR (kg/d) = 2, BW (kg) = 50, EF = 1, are considered. HQ are summarized in Table 8. Ed (years) = 7, Total Period (years) = 33 For carcinogenic effects: the excess risk of cancer Province Yatenga Passoré Zondoma Lorom from exposure to a chemical is described in terms of the C (µg/L) 10 16 10 87 10 15 0 0 probability that an exposed individual will develop can- ED (μg·kg−1·d−1) 0.4 0.64 0.4 3.48 0.4 0.6 0 0 cer because of that exposure by age 70. For each chemical of concern, this value is calculated from the daily

intake of the chemical from the site averaged over a life- time (DIL) and the slope factor (SF) for the chemical, as follows: Excess Cancer Risk=1-exp -DIL SF (4) In most cases (except when the product of DIL × SF is larger than about 0.01), this equation may be accurately approximated by the following: Excess Cancer Risk DIL SF (5) Excess cancer risks are summed across all chemicals of concern and all exposure pathways that contribute to exposure of an individual in a given population. Results from our study (Individual Excess Cancer Risk, quoted

IECR) are in Table 9. Figure 2. Skin hyper pigmentation. The level of total cancer risk that is of concern is a matter of personal, community, and regulatory judgment shows the target exposure and the effects of arsenic on [14]. In our study, the risk induced by arsenic for non- humans in the North region, Burkina Faso. cancer effects is acceptable when HQ is less than 1E + In summary, chronic exposure to arsenic is known as 00.When children aged 0 - 14 are subject to Arsenic arsenicosis and there is a significant latency period be- concentrations below 87 g/L the risk is huge in all cases fore symptoms are developed. There appears to be dis- and HQ > 1. These concentrations can cause disease to crepancy in the literature regarding latency, with some these children. However, concentrations under the norm reports of 2 years being the minimum for hyper pigmen- can induce the same disease when other environmental tation and keratosis. Researchers in Bangladesh suggest factors are taken into account. Adults (from 15 years) that 5 years is the minimum latency, whilst some other subject to As pollution have a high risk of getting sick estimates suggest that this is 9 years. Latency for cancers when the arsenic concentration exceeds the standard of 10 is also unknown, but it is estimated to be of the order of μg/L. The risk is increased when other environmental 20 years [15]. conditions are taken into account. For carcinogenic effects, the World Health Organiza- 3.2. Risk Characterization tion (WHO) gives as limit HQ of 10−5 [17] which is far For risk characterization, we follow the protocol de- below calculated values in our study. Individuals in the scribed by the United States Environmental Protection region are exposed to arsenic concentrations higher than Agency [16]. 10 μg/L and are at very high risk of having cancer because Non carcinogenic effects: the potential for non-cancer values HQ that we find are more than 60 times the WHO effects is evaluated by comparing the estimated daily guide. Water containing As values above the guideline

Table 6. Calculated daily intake in the North region.

Ed = 33, Q (kg/d) = 2 , BW = 65 kg, TE = 1, Total Period = 52 Province Yatenga Passoré Zondoma Loroum C (µg/L) 5 10 16 5 10 86 5 10 16 5 5 5 DI (μg·kg−1·d−1) 0.0976 0.195 0.312 0.0976 0.195 1.68 0.9976 0.195 0.312 0.0976 0.0976 0.0976

Table 7. Hazard Quotient (HQ) for non-carcinogenic effects in Scenario 1.

Scénario 1 HQ = DI/RfD; Acceptable risk if HQ < 1 Population now exposed, individuals from 0 to 14 years, RfD (VTR = 7×10−5 mg·kg−1·d−1) Yatenga Passore Zondoma Loroum DI (μg·kg−1·d−1) 0.34 0.55 0.34 3 0.34 0.51 0 0 Hazard quotient (HQ) 4.92 7.88 4.92 42.85 4.92 7.38 0 0

Table 8. Hazard Quotient (HQ) for non-carcinogenic effects in Scenario 2.

Scénario 1 HQ =DI/RfD; Acceptable risk if HQ < 1 Population now exposed, individuals from 15 to 70 years, RfD (VTR = 7×10−5 mg·kg−1·d−1) Yatenga Passore Zondoma Loroum DI (μg·kg−1·d−1) 0.34 0.55 0.34 3 0.34 0.51 0 0 Hazard quatient (HQ) 5.71 9.14 5.71 49.71 5.71 8.51 0 0

Table 9. Calculated Individual Excess Cancer Risk (IECR) and Collective Excess Cancer Risk (CECR).

Scenario 1: Individuals from 15 to 70 years

Unitary Excess Risk (UER = 1.5), IECR = DI × UER, CECR = UICR × exposed population

Province Yatenga Passoré Zondoma Loroum ED (μg·kg−1·d−1) 0.0976 0.195 0.312 0.0976 0.195 1.68 0.9976 0.195 0.312 0.0976 0.0976 0.0976 IECR (×10−4) 1.46 2.93 4.69 1.46 2.93 25.2 1.46 2.93 4.69 1.46 1.46 1.46 CECR 67 133 213 39 78 345 20 40 55 17 17 17

of non carcinogenic (Figure 2) and carcinogenic (Figure 3) risks to which people are exposed in the North region.

4. Conclusions The delayed effects on health of exposure to arsenic, the lack of common definitions, the lack of local awareness and the poor reporting in affected areas are the main ob- stacles that stumble determining the magnitude of the problem of arsenic in drinking water in Burkina Faso. Clinical cases of cancers associated with arsenic in drinking water were reported in the provinces of Passoré, Zondoma and Yatenga. This study allowed us to assess the extent of problems related to arsenic in groundwater. Pollution of groundwater caused the health risk and death Figure 3. Palmar keratosis (skin cancer). in the North region. Under the light of the results this study, it seems clear value should therefore be subjected to prior treatment. that measures should be taken at various levels to try to Affordable technologies are available [18]. When this address the risks associated with water contamination by treatment is not possible these water points should be arsenic. Firstly appropriate technologies for arsenic re- prohibited for human consumption until treatment is moval from drinking water should be considered and found. secondly public targeted information should be devel- The following photos illustrate reported clinical cases oped in order to confront arsenic with appropriate tech-

Copyright © 2013 SciRes. JWARP 52 J. F. NZIHOU ET AL. nologies rather than fly it by closing water points. [9] M. Rahman, V. M. Sohel, M. Yunus, M. A. Wahed and P. K. Stretfrield, “Arsenic Exposure and Age-Sex Specific Risk for Skin Lesions: A Population Based Case-Referent REFERENCES Study in Bangladesh,” Environ Health Perspect, Vol. 114, No. 12, 2006, pp. 1847-1852. [1] Ministry of Agriculture and Water Resources, “2011 Sta- tistical Yearbook of Drinking Water and Wastewater and [10] INERIS, “Choix des Valeurs Toxicologiquesde Référence Excreta,” 2011, pp. 35-40. (VTR)-Arsenic,” 2006. http://cns.bf/IMG/pdf/meaha_annuairestatistiqueaepa_20 http://www.ineris.fr/centredoc/06DR082__rapport_final.p 11_vf-2.pdf df [2] Water and Environment Program of the North, “Rea- [11] W. P. Tseng, H. M. Chu, S. W. How, J. M. Fong, C. S. lization of 360 Drillings in Villages of the Provinces of Lin and S. Yeh, “Prevalence of Skin Cancer in an En- Yatenga, Lorum, Passoré and Zondoma from 1994 to demic Area of Chronic Arsenicism in Taiwan,” Journal 2004,” Arsenic Poisoning in the North: Two Dead and of the National Cancer Institute, Vol. 40, No. 3, 1968, pp. Eleven Wells Embargoed. 453-463. http://www.lefaso.net/spip.php?article18173 [12] W. P. Tseng, “Effects and Dose-Response Relationships [3] P. L. Smedley, J. Knudsen and D. Maiga, “Arsenic in of Skin Cancer and Blackfoot Disease with Arsenic,” Groundwater from Mineralised Proterozoic Basement Environmental Health Perspectives, Vol. 19, 1977, pp 109- Rocks of Burkina Faso,” Applied Geochemistry, Vol. 22, 119. doi:10.1289/ehp.7719109 No. 5, 2007, pp. 1074-1092. [13] Institut National de la Statistique et de la Démographie, doi:10.1016/j.apgeochem.2007.01.001 “Résultats Définitifs du Recensement Général de la Po- [4] I. T. Somé, A. K. Sakira, M. Ouédraogo, T. Z. Ouédraogo, pulation et de l’Habitat 2006,” 2009. A. Traoré, B. Sondo and P. I. Guissou, “Arsenic Levels in http://cns.bf/IMG/pdf/population_1997-2006.pdf Tube-Wells Water, Food, Residents’ Urine and the Preva- [14] Agency for Toxic Substances & Disease Registry, “Pub- lence of Skin Lesions in Yatenga Province, Burkina lic Health Assessment Guidance Manual. Appendix G: Baso,” Interdisciplinary Toxicology, Vol. 5, No. 1, 2012, Calculating Exposure Doses,” 2005. pp. 38-41. http://www.atsdr.cdc.gov/hac/phamanual/appg.html http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3389508 [15] World Health Organization, “Arsenic, Drinking-Water and [5] Institut Géographique du Burkina, “Cartoweb-Region Nord,” Health Risk Substitution in Arsenic Mitigation: A Dis- 2012. http://www.igb.bf/pagehtm/carteweb.htm cussion Paper,” 2012. [6] S. Nakolendousse, “Méthode d’Evaluation de la Produc- http://www.who.int/water_sanitation_health/dwq/wsh030 tivité des Sites Aquifères au Burkina Faso,” Ph.D. The- 6/en/index2.html sis, Université Joseph Fourier, Grenoble, 1991. [16] United States Environmental Protection Agency, “Risk [7] P. L. Smedley and D. G. Kinniburgh, “A Review of the Characterization,” 2013. Source, Behaviour and Distribution of Arsenic in Natural http://www.epa.gov/region8/r8risk/hh_risk.html Waters,” Applied Geochemistry, Vol. 17, No. 5, 2002 pp. [17] World Health Organization, “Drinking Water Guidelines 517-568. and Standards,” 2013. [8] D. van Halem, S. A. Bakker, G. L. Amy and J. C. van http://www.who.int/water_sanitation_health/dwq/arsenicu Dijk, “Arsenic in Drinking Water: A Worldwide Water n5.pdf Quality Concern for Water Supply Companies,” Drinking [18] N. R. Khandaker, P. V. Brady and J. L. Krumhansl, “Ar- Water Engineering Science, Vol. 2, No. 1, 2009, pp. 29- senic Removal from Drinking Water: A Handbook for 34. Communities,” http://www.drink-water-eng-sci.net/2/29/2009/dwes-2-29 http://www.sandia.gov/water/docs/NRK_PVB_JLK_As_ -2009.pdf Book.pdf